The heritability of alcoholism is 40-60% in both men and women however, as in other complex psychiatric diseases it has proved difficult to identify causative genes. Intermediate phenotypes are associated biological traits that may be influenced by variation at fewer genes and may mediate different aspects of the disease. The intermediate phenotypes for alcoholism that we are studying include dimensional anxiety (harm avoidance (HA)), resting EEG phenotypes, event-related potentials (ERPs)and heart rate variability (HRV). We have three large intermediate phenotype datasets: 247 U.S. Caucasians, 365 Plains American Indians with a high prevalence of alcoholism and 198 Southeastern American Indians with a low prevalence of alcoholism. In collaboration with Dr Zubietta from the University of Michigan we are also studying imaging data (fMRI) as an intermediate phenotype for addiction-related phenotypes. We have recently performed a dense whole genome linkage scan with 3878 unlinked SNPs spaced at an average distance of 1cM on the Plains Indian sample. These Plains Indians derive from six pedigrees, the largest of which includes 1004 individuals. Linkage analysis was performed using SOLAR. There was no significant linkage with alcoholism in the Plains Indians. We then looked for linkage with resting EEG power, an intermediate phenotype for alcoholism. We found that EEG power was stable over time and moderately heritable across all frequency bands (0.27 to 0.58). There was a convergence of linkage peaks for alpha, beta and theta EEG power on chromosome (chr) 5 with LOD scores of 3.5. The gene for corticotropin releasing hormone binding protein (CRHBP) was located at the apex of the convergent linkage peaks. CRHBP is implicated in stress and addiction. Subsequent analyses showed that CRHBP SNPs and haplotypes were significantly associated with alpha EEG power in the Plains Indians and also the U.S. Caucasians. Moreover, the same CRHBP SNPs were significantly associated with anxiety disorders in the Plains Indians and alcohol use disorders in the U.S. Caucasians (Enoch et al., 2008). These results suggest a likely role for CRHBP in stress-related disorders including alcoholism. Indeed, we have recently shown that in a sample of African American men with substance dependence, the same distal CRHBP SNPs predicted suicidal behavior in those individuals who had been exposed to severe childhood trauma (Roy et al, submitted). CRHBP is buffered from adjacent genes by several haplotype blocks indicating that a functional locus is likely to reside within this gene or its environs. We have demonstrated the presence of an alternative CRHBP isoform in brain in which the terminal exon is spliced out in favor of two alternative exons resulting in a change in peptide sequence that might affect protein folding and stability resulting in altered CRH binding affinity. Functional studies are currently underway. A linkage peak for theta EEG power with a LOD score of 3.2 was found on chr 22. There were suggestive peaks (LOD = 2.2 2.5) on chrs 4, 10 and 11. There are three candidate genes at the apex of the linkage peak on chr 11: serotonin receptors 3A and 3B (HTR3A and HTR3B) and DRD2. We found a significant association between HTR3B and EEG alpha power in both the Plains Indians and U.S. Caucasians and with antisocial alcoholism in Finnish Caucasians (Ducci et al.,2009). Moreover, in a sample of African American men with substance abuse, we found that a functional HTR3B SNP rs1176744 was associated with alcohol dependence (Enoch et al, in press, Mol Psychiatry, 2010). We will be looking at other candidate genes under the linkage peaks in the near future. Our studies have shown that intermediate phenotypes are particularly useful for identifying genes for alcoholism in populations with a high prevalence of this disease. Recent collaborative studies with Dr Zubietta have revealed that genetic variation in the oxytocin gene is associated with fMRI changes (activation) in the nucleus accumbens during the anticipation of reward, a marker for vulnerability to addiction. Also, variation in the CRHR1 gene that encodes the stress-response CRH1 receptor was associated with differences in brain activity in response to emotional stimuli. Thus these results implicate brain mechanisms by which variation in CRHR1 may contribute to differences in emotional response to stress and thereby vulnerability to alcoholism and other stress-related disorders. Formerly titled "Genetic studies of the electroencephalogram and event-related potentials" and "Genetic studies of EEG and ERP traits related to alcoholism".